9

Radiological aspects

T he objective of this chapter is to provide criteria with which to assess the safety of
drinking-water with respect to its radionuclide content. The Guidelines do not
differentiate between naturally occurring and artificial or human-made radionuclides.
The guidance values for radioactivity in drinking-water recommended in the first
edition of the Guidelines were based on the risks of exposure to radiation sources and
the health consequences of exposure to radiation. The second edition of the Guidelines
incorporated the 1990 recommendations of the International Commission on Radio-
logical Protection (ICRP, 1991). The third edition incorporates recent developments,
including the ICRP publications on prolonged exposures and on dose coefficients.
Radiological hazards may derive from ionizing radiation emitted by a number of
radioactive substances (chemicals) in drinking-water. Such hazards from drinking-
water are rarely of public health significance, and radiation exposure from drinking-
water must be assessed alongside exposure from other sources.
The approach taken in the Guidelines for controlling radiological hazards has two
stages:
— initial screening for gross alpha and/or beta activity to determine whether the
activity concentrations (in Bq/litre) are below levels at which no further action
is required; and
— if these screening levels are exceeded, investigation of the concentrations of indi-
vidual radionuclides and comparison with specific guidance levels.
The risk due to radon in drinking-water derived from groundwater is typically low
compared with that due to total inhaled radon but is distinct, as exposure occurs
through both consumption of dissolved gas and inhalation of released radon and its
daughter radionuclides. Greatest exposure is general ambient inhalation and inhala-
tion from terrestrial sources, where the gas is infiltrating into dwellings, especially into
basements. Radon of groundwater origin would usually be a small increment of the
total, but may indicate deposits in the region that are emitting into basements.
The screening and guidance levels apply to routine (“normal”) operational condi-
tions of existing or new drinking-water supplies. They do not apply to a water supply

197
 GUIDELINES FOR DRINKING-WATER QUALITY

contaminated during an emergency involving the release of radionuclides into the

environment. Guidance and generic action levels covering emergency situations are
available elsewhere (IAEA, 1996, 1997, 1999, 2002).
The current Guidelines are based on:

— a recommended reference dose level (RDL) of the committed effective dose, equal
to 0.1 mSv from 1 year’s consumption of drinking-water (from the possible total
radioactive contamination of the annual drinking-water consumption). This
comprises 10% of the intervention exemption level recommended by the ICRP
for dominant commodities (e.g., food and drinking-water) for prolonged expo-
sure situations, which is most relevant to long-term consumption of drinking-
water by the public (ICRP, 2000). The RDL of 0.1 mSv is also equal to 10% of the
dose limit for members of the population, recommended by both the ICRP (1991)
and the International Basic Safety Standards (IAEA, 1996). These are accepted by
most WHO Member States, the European Commission, FAO and WHO.
— dose coefficients for adults, provided by the ICRP.

The additional risk to health from exposure to an annual dose of 0.1 mSv associated
with the intake of radionuclides from drinking-water is considered to be low for the
following reasons:

• The nominal probability coefficient for radiation-induced stochastic health effects,

which include fatal cancer, non-fatal cancer and severe hereditary effects for the
whole population, is 7.3 ¥ 10-2/Sv (ICRP, 1991). Multiplying this by an RDL equal
to 0.1 mSv annual exposure via drinking-water gives an estimated upper-bound
lifetime risk of stochastic health effects of approximately 10-4, which can be con-
sidered small in comparison with many other health risks. This reference risk esti-
mation for radionuclides is quite reliable due to the extensive scientific databases
that have included human population exposure data. As with chemical carcinogen
risk extrapolations, the lower-bound risk is zero.
• Background radiation exposures vary widely across the Earth, but the average is
about 2.4 mSv/year, with the highest local levels being up to 10 times higher without
any detected increased health risks from population studies; 0.1 mSv therefore
represents a small addition to background levels.

9.1 Sources and health effects of radiation exposure

Environmental radiation originates from a number of naturally occurring and
human-made sources. Radioactive materials occur naturally everywhere in the en-
vironment (e.g., uranium, thorium and potassium-40). By far the largest proportion
of human exposure to radiation comes from natural sources – from external sources
of radiation, including cosmic and terrestrial radiation, and from inhalation or inges-
tion of radioactive materials (Figure 9.1). The United Nations Scientific Committee

198
 9. RADIOLOGICAL ASPECTS

Radon (natural
internal exposure)
Other artificial 43%
(human-made)
sources
1%

Food, water
(natural internal
exposure)
8%
Earth gamma
radiation
(natural external
exposure)
15%

Medical exposure
20%

Cosmic rays
(natural external exposure)
13%

Figure 9.1 Sources and distribution of average radiation exposure for the world population

on the Effects of Atomic Radiation (UNSCEAR, 2000) has estimated that the global
average annual human exposure from natural sources is 2.4 mSv/year (Table 9.1).
Some sources (e.g., uranium) can be concentrated during extraction by mining and
other industrial activities.
There are large local variations in human exposure to radiation, depending on a
number of factors, such as height above sea level, the amount and type of radio-
nuclides in the soil (terrestrial exposure), the composition of radionuclides in the air,
food and drinking-water and the amount taken into the body via inhalation or inges-
tion. There are certain areas of the world, such as parts of the Kerala state in India
and the Pocos del Caldas plateau in Brazil, where levels of background radiation are

Table 9.1 Average radiation dose from natural sources

Worldwide average
annual effective dose Typical range
Source (mSv) (mSv)
External exposure
Cosmic rays 0.4 0.3–1.0
Terrestrial gamma raysa 0.5 0.3–0.6
Internal exposure
Inhalation (mainly radon) 1.2 0.2–10b
Ingestion (food and drinking-water) 0.3 0.2–0.8
Total 2.4 1–10
a
Terrestrial exposure is due to radionuclides in the soil and building materials.
b
Dose from inhalation of radon may exceed 10 mSv/year in certain residential areas.
Source: UNSCEAR (2000).

199
 GUIDELINES FOR DRINKING-WATER QUALITY

relatively high. Levels of exposure for the general population in such areas may be up
to 10 times higher than the average background level of 2.4 mSv given in Table 9.1.
No deleterious health effects associated with this elevated radiation exposure have
been detected (UNSCEAR, 2000).
Several radioactive compounds may be released into the environment, and hence
into drinking-water supplies, from human activities and human-made sources (e.g.,
from medical or industrial use of radioactive sources). The worldwide per capita effec-
tive dose from diagnostic medical examination in 2000 was 0.4 mSv/year (typical
range is 0.04–1.0 mSv/year, depending on level of health care). There is only a very
small worldwide contribution from nuclear power production and nuclear weapons
testing. The worldwide annual per capita effective dose from nuclear weapons testing
in 2000 was estimated at 0.005 mSv; from the Chernobyl accident, 0.002 mSv; and
from nuclear power production, 0.0002 mSv (UNSCEAR, 2000).

9.1.1 Radiation exposure through drinking-water

Radioactive constituents of drinking-water can result from:
— naturally occurring radioactive species (e.g., radionuclides of the thorium and
uranium decay series in drinking-water sources), in particular radium-226/228
and a few others;
— technological processes involving naturally occurring radioactive materials (e.g.,
the mining and processing of mineral sands or phosphate fertilizer production);
— radionuclides discharged from nuclear fuel cycle facilities;
— manufactured radionuclides (produced and used in unsealed form), which
might enter drinking-water supplies as a result of regular discharges and, in par-
ticular, in case of improper medical or industrial use and disposal of radioac-
tive materials; such incidents are different from emergencies, which are outside
the scope of these Guidelines; and
— past releases of radionuclides into the environment, including water sources.
The contribution of drinking-water to total exposure is typically very small and is
due largely to naturally occurring radionuclides in the uranium and thorium decay
series. Radionuclides from the nuclear fuel cycle and from medical and other uses of
radioactive materials may, however, enter drinking-water supplies. The contributions
from these sources are normally limited by regulatory control of the source or practice,
and it is normally through this regulatory mechanism that remedial action should be
taken in the event that such sources cause concern by contaminating drinking-water.

9.1.2 Radiation-induced health effects through drinking-water

There is evidence from both human and animal studies that radiation exposure at low
to moderate doses may increase the long-term incidence of cancer. Animal studies in
particular suggest that the rate of genetic malformations may be increased by radia-
tion exposure.

200
 9. RADIOLOGICAL ASPECTS

No deleterious radiological health effects are expected from consumption of

drinking-water if the concentrations of radionuclides are below the guidance levels
(equivalent to a committed effective dose below 0.1 mSv/year).
Acute health effects of radiation, leading to reduced blood cell counts and, in very
severe cases, death, occur at very high doses of exposure of the whole body or large
part of the body (IAEA, 1998). Due to the low levels of radionuclides typically found
in drinking-water supplies, acute health effects of radiation are not a concern for
drinking-water supplies.

9.2 Units of radioactivity and radiation dose

The SI unit of radioactivity is the becquerel (Bq), where 1 Bq = 1 disintegration per
second. Guidance levels for drinking-water are given as the activity of the radio-
nuclide per litre, called the activity concentration (Bq/litre). The radiation dose
resulting from ingestion of a radionuclide depends on a number of chemical and
biological factors. These include the fraction of the intake that is absorbed from the
gut, the organs or tissues to which the radionuclide is transported and the time during
which the radionuclide remains in the organ or tissue before excretion. The nature of
the radiation emitted on decay and the sensitivity of the irradiated organs or tissues
to radiation must also be considered.
The absorbed dose refers to how much energy is deposited in material by the radi-
ation. The SI unit for absorbed dose is the gray (Gy), where 1 Gy = 1 J/kg (joule per
kilogram).
The equivalent dose is the product of the absorbed dose and a factor related to the
particular type of radiation (depending on the ionizing capacity and density).
The effective dose of radiation received by a person is, in simple terms, the sum of
the equivalent doses received by all tissues or organs, weighted for “tissue weighting
factors.” These reflect different sensitivities to radiation of different organs and tissues
in the human body. The SI unit for the equivalent and effective dose is the sievert (Sv),
where 1 Sv = 1 J/kg.
To reflect the persistence of radionuclides in the body once ingested, the committed
effective dose is a measure of the total effective dose received over a lifetime (70 years)
following intake of a radionuclide (internal exposure).
The term “dose” may be used as a general term to mean either absorbed dose (Gy)
or effective dose (Sv), depending on the situation. For monitoring purposes, doses are
determined from the activity concentration of the radionuclide in a given material.
In the case of water, activity concentration is given in becquerels per litre (Bq/litre).
This value can be related to an effective dose per year (mSv/year) using a dose coef-
ficient (mSv/Bq) and the average annual consumption of water (litres/year).
The effective dose arising from the ingestion of a radioisotope in a particular
chemical form can be estimated using a dose coefficient. Data for age-related dose
coefficients for ingestion of radionuclides have been published by the ICRP and the
International Atomic Energy Agency (IAEA). Table 9.2 shows the dose coefficients for

201
 GUIDELINES FOR DRINKING-WATER QUALITY

Table 9.2 Dose coefficients for ingestion of radionuclides by adult members of the public
Category Radionuclide Dose coefficient (mSv/Bq)
Natural uranium series Uranium-238 4.5 ¥ 10-5
Uranium-234 4.9 ¥ 10-5
Thorium-230 2.1 ¥ 10-4
Radium-226 2.8 ¥ 10-4
Lead-210 6.9 ¥ 10-4
Polonium-210 1.2 ¥ 10-3
Natural thorium series Thorium-232 2.3 ¥ 10-4
Radium-228 6.9 ¥ 10-4
Thorium-228 7.2 ¥ 10-5
Fission products Caesium-134 1.9 ¥ 10-5
Caesium-137 1.3 ¥ 10-5
Strontium-90 2.8 ¥ 10-5
Iodine-131 2.2 ¥ 10-5
Other radionuclides Tritium 1.8 ¥ 10-8
Carbon-14 5.8 ¥ 10-7
Plutonium-239 2.5 ¥ 10-4
Americium-241 2.0 ¥ 10-4

naturally occurring radionuclides or those arising from human activities that might
be found in drinking-water supplies (IAEA, 1996; ICRP, 1996).

9.3 Guidance levels for radionuclides in drinking-water

The guidance levels for radionuclides in drinking-water are presented in Table 9.3 for
radionuclides originating from natural sources or discharged into the environment as
the result of current or past activities. These levels also apply to radionuclides released
due to nuclear accidents that occurred more than 1 year previously. The activity con-
centration values in Table 9.3 correspond to an RDL of 0.1 mSv/year from each
radionuclide listed if their concentration in the drinking-water consumed during the
year does not exceed these values. The associated risk estimate was given at the begin-
ning of this chapter. However, for the first year immediately after an accident, generic
action levels for foodstuffs apply as described in the International Basic Safety
Standards (IAEA, 1996) and other relevant WHO and IAEA publications (WHO,
1988; IAEA, 1997, 1999).
The guidance levels for radionuclides in drinking-water were calculated by the
following equation:
GL = IDC (h ing◊q )
where:
GL = guidance level of radionuclide in drinking-water (Bq/litre),
IDC = individual dose criterion, equal to 0.1 mSv/year for this calculation,
hing = dose coefficient for ingestion by adults (mSv/Bq),
q = annual ingested volume of drinking-water, assumed to be 730 litres/year.

202
 9. RADIOLOGICAL ASPECTS

Table 9.3 Guidance levels for radionuclides in drinking-water

Guidance Guidance Guidance
level level level
Radionuclides (Bq/litre)a Radionuclides (Bq/litre)a Radionuclides (Bq/litre)a
3 97m 149
H 10 000 Tc 100 Pm 100
7 99 151
Be 10 000 Tc 100 Sm 1000
14 97 153
C 100 Ru 1000 Sm 100
22 103 152
Na 100 Ru 100 Eu 100
32 106 154
P 100 Ru 10 Eu 100
33 105 155
P 1 000 Rh 1000 Eu 1000
35 103 153
S 100 Pd 1000 Gd 1000
36 105 160
Cl 100 Ag 100 Tb 100
45 110m 169
Ca 100 Ag 100 Er 1000
47 111 171
Ca 100 Ag 100 Tm 1000
46 109 175
Sc 100 Cd 100 Yb 1000
47 115 182
Sc 100 Cd 100 Ta 100
48 115m 181
Sc 100 Cd 100 W 1000
48 111 185
V 100 In 1000 W 1000
51 114m 186
Cr 10 000 In 100 Re 100
52 113 185
Mn 100 Sn 100 Os 100
53 125 191
Mn 10 000 Sn 100 Os 100
54 122 193
Mn 100 Sb 100 Os 100
55 124 190
Fe 1 000 Sb 100 Ir 100
59 125 192
Fe 100 Sb 100 Ir 100
56 123m 191
Co 100 Te 100 Pt 1000
57 127 193m
Co 1 000 Te 1000 Pt 1000
58 127m 198
Co 100 Te 100 Au 100
60 129 199
Co 100 Te 1000 Au 1000
59 129m 197
Ni 1 000 Te 100 Hg 1000
63 131 203
Ni 1 000 Te 1000 Hg 100
65 131m 200
Zn 100 Te 100 Tl 1000
71 132 201
Ge 10 000 Te 100 Tl 1000
73 125 202
As 1 000 I 10 Tl 1000
74 126 204
As 100 I 10 Tl 100
76 129 203
As 100 I 1 Pb 1000
77 131 210
As 1 000 I 10 Pbb 0.1
75 129 206
Se 100 Cs 1000 Bi 100
82 131 207
Br 100 Cs 1000 Bi 100
86 132 210 b
Rb 100 Cs 100 Bi 100
85 134 210
Sr 100 Cs 10 Pob 0.1
89 135 223
Sr 100 Cs 100 Rab 1
90 136 224
Sr 10 Cs 100 Rab 1
90 137 225
Y 100 Cs 10 Ra 1
91 131 226
Y 100 Ba 1000 Rab 1
93 140 228
Zr 100 Ba 100 Rab 0.1
95 140 227
Zr 100 La 100 Thb 10
93m 139 228
Nb 1 000 Ce 1000 Thb 1
94 141 229
Nb 100 Ce 100 Th 0.1
95 143 230
Nb 100 Ce 100 Thb 1
93 144 231
Mo 100 Ce 10 Thb 1000
99 143 232
Mo 100 Pr 100 Thb 1
96 147 234
Tc 100 Nd 100 Thb 100
97 147 230
Tc 1000 Pm 1000 Pa 100

continued

203
 GUIDELINES FOR DRINKING-WATER QUALITY

Table 9.3 Continued

Guidance Guidance Guidance
level level level
Radionuclides (Bq/litre) Radionuclides (Bq/litre) Radionuclides (Bq/litre)
231 b 238 247
Pa 0.1 Pu 1 Cm 1
233 239 248
Pa 100 Pu 1 Cm 0.1
230 240 249
U 1 Pu 1 Bk 100
231 241 246
U 1 000 Pu 10 Cf 100
232 242 248
U 1 Pu 1 Cf 10
233 244 249
U 1 Pu 1 Cf 1
234 b 241 250
U 1 Am 1 Cf 1
235 b 242 251
U 1 Am 1000 Cf 1
236 b 242m 252
U 1 Am 1 Cf 1
237 243 253
U 100 Am 1 Cf 100
238 b,c 242 254
U 10 Cm 10 Cf 1
237 243 253
Np 1 Cm 1 Es 10
239 244 254
Np 100 Cm 1 Es 10
236 245 254m
Pu 1 Cm 1 Es 100
237 246
Pu 1000 Cm 1
a
Guidance levels are rounded according to averaging the log scale values (to 10n if the calculated value was below
3 ¥ 10n and above 3 ¥ 10n-1).
b
Natural radionuclides.
c
The provisional guideline value for uranium in drinking-water is 15 mg/litre based on its chemical toxicity for the
kidney (see section 8.5).

The higher age-dependent dose coefficients calculated for children (accounting for
the higher uptake and/or metabolic rates) do not lead to significantly higher doses
due to the lower mean volume of drinking-water consumed by infants and children.
Consequently, the recommended RDL of committed effective dose of 0.1 mSv/year
from 1 year’s consumption of drinking-water applies independently of age.

9.4 Monitoring and assessment for dissolved radionuclides

9.4.1 Screening of drinking-water supplies
The process of identifying individual radioactive species and determining their con-
centration requires sophisticated and expensive analysis, which is normally not justi-
fied, because the concentrations of radionuclides in most circumstances are very low.
A more practical approach is to use a screening procedure, where the total radioac-
tivity present in the form of alpha and beta radiation is first determined, without
regard to the identity of specific radionuclides.
Screening levels for drinking-water below which no further action is required are
0.5 Bq/litre for gross alpha activity and 1 Bq/litre for gross beta activity. The gross beta
activity screening level was published in the second edition of the Guidelines and,
in the worse case (radium-222), would lead to a dose close to the guidance RDL of
0.1 mSv/year. The screening level for gross alpha activity is 0.5 Bq/litre (instead of
the former 0.1 Bq/litre), as this activity concentration reflects values nearer the
radionuclide-specific guidance RDL.

204
 9. RADIOLOGICAL ASPECTS

9.4.2 Strategy for assessing drinking-water

If either of the screening levels is exceeded, then the specific radionuclides producing
this activity should be identified and their individual activity concentrations meas-
ured. From these data, an estimate of committed effective dose for each radionuclide
should be made and the sum of these doses determined. If the following additive
formula is satisfied, no further action is required:

Ci
 GL i
£1
i

where:
Ci = the measured activity concentration of radionuclide i, and
GLi = the guidance level value (see Table 9.3) of radionuclide i that, at an intake of
2 litres/day for 1 year, will result in a committed effective dose of 0.1 mSv/year.
Where the sum exceeds unity for a single sample, the RDL of 0.1 mSv would be
exceeded only if the exposure to the same measured concentrations were to continue
for a full year. Hence, such a sample does not in itself imply that the water is unsuitable
for consumption but should be regarded as an indication that further investigation,
including additional sampling, is needed. Gross beta and gross alpha activity screen-
ing has to be repeated first, then radionuclide-specific analysis conducted only if sub-
sequently measured gross values exceed the recommended practical screening values
(1 Bq/litre and 0.5 Bq/litre, respectively).
The application of these recommendations is summarized in Figure 9.2.
The gross beta measurement includes a contribution from potassium-40, a beta
emitter that occurs naturally in a fixed ratio to stable potassium. Potassium is an essen-
tial element for humans and is absorbed mainly from ingested food. Potassium-40
does not accumulate in the body but is maintained at a constant level independent of
intake. The contribution of potassium-40 to beta activity should therefore be sub-
tracted following a separate determination of total potassium. The specific activity of
potassium-40 is 30.7 Bq/g of potassium. However, not all the radiation from
potassium-40 appears as beta activity. The beta activity of potassium-40 is 27.6 Bq/g
of stable potassium, which is the factor that should be used to calculate the beta
activity due to potassium-40.

9.4.3 Remedial measures

If the RDL of 0.1 mSv/year is being exceeded on aggregate, then the options available
to the competent authority to reduce the dose should be examined. Where remedial
measures are contemplated, any strategy considered should first be justified (in the
sense that it achieves a net benefit) and then optimized in accordance with the
recommendations of ICRP (1989, 1991) in order to produce the maximum net
benefit.

205
 GUIDELINES FOR DRINKING-WATER QUALITY

Determine gross α
and gross β activity

Gross α ≤ 0.5 Bq/litre Gross α > 0.5 Bq/litre

and or
Gross β ≤ 1 Bq/litre Gross β > 1 Bq/litre

Determine individual
radionuclide concentrations
and compare with
guidance levels

Dose ≤ 0.1 mSv Dose > 0.1 mSv

Water suitable, Consider and,

no further action when justified, take
necessary remedial action to
reduce dose

Figure 9.2 Application of screening and guidance levels for radionuclides in drinking-water

9.5 Radon
9.5.1 Radon in air and water
The largest fraction of natural radiation exposure comes from radon, a radioactive
gas (see Table 9.1 and Figure 9.1), due to decay of radium contained in rocks and soil
as part of the uranium radionuclide chain. The term radon in general refers mostly
to radon-222. Radon is present virtually everywhere on Earth, but particularly in the
air over land and in buildings.
Underground rock containing natural uranium continuously releases radon into
water in contact with it (groundwater). Radon is readily released from surface water;
consequently, groundwater has potentially much higher concentrations of radon than
surface water. The average concentration of radon is usually less than 0.4 Bq/litre in
public water supplies derived from surface waters and about 20 Bq/litre from ground-
water sources. However, some wells have been identified with higher concentrations,
up to 400 times the average, and in rare cases exceeding 10 kBq/litre.
In assessing the dose from radon ingestion, it is important that water processing
technology that can remove radon be considered before consumption is taken into
account. Moreover, the use of radon-containing groundwater supplies not treated for
radon removal (usually by aeration) for general domestic purposes will increase the
levels of radon in the indoor air, thus increasing the dose from indoor inhalation. This

206
 GUIDELINES FOR DRINKING-WATER QUALITY

dose depends markedly on the forms of domestic usage and housing construction
(NCRP, 1989), because most of the indoor air radon usually enters from the founda-
tion of the house in contact with the ground rather than from the water. The amount

206a
 9. RADIOLOGICAL ASPECTS

and form of water intake, other domestic usage of water and the construction of
houses vary widely throughout the world.
UNSCEAR (2000) refers to a US NAS (1999) report and calculates the “average
doses from radon in drinking water to be as low as 0.025 mSv/year via inhalation and
0.002 mSv/year from ingestion” compared with the inhalation dose of 1.1 mSv/year
from radon and its decay products in air.

9.5.2 Risk
Large pooled studies of indoor radon and lung cancer risk have recently become avail-
able. The European pooled analysis of 13 indoor radon studies estimated a 16% risk
increase per 100 Bq/m3 (Darby et al., 2005). Based on these data, radon accounts for
about 9% of all lung cancer deaths and 2% of total cancer deaths in Europe. Similar
results were obtained from the joint analysis of North American radon studies
(Krewski et al., 2005).
For the USA, the US EPA has estimated that radon causes about 21 000 lung cancer
deaths per year (with an uncertainty range of 8000–45 000), out of about 160 000
annual lung cancer deaths (US EPA, 2003). Radon is the second leading cause of lung
cancer, after smoking.
US NAS (1999) reports an approximately 100-fold smaller risk from exposure to
radon in drinking-water (i.e., 183 deaths each year). In addition to the 19 000 deaths
from lung cancer caused by radon in indoor air, a further 160 were estimated to result
from inhaling radon that was emitted from water used in the home. For comparison,
about 700 lung cancer deaths each year were attributed to exposure to natural levels
of radon while people are outdoors.
The US NAS (1999) also assessed that the risk of stomach cancer caused by
drinking-water that contains dissolved radon is extremely small, with the probability
of about 20 deaths annually compared with the 13 000 deaths from stomach cancer
that arise each year from other causes in the USA.

9.5.3 Guidance on radon in drinking-water supplies

Controls should be implemented if the radon concentration of drinking-water for
public water supplies exceeds 100 Bq/litre. Any new, especially public, drinking-water
supply using groundwater should be tested prior to being used for general consump-
tion. If the radon concentration exceeds 100 Bq/litre, treatment of the water source
(see section 9.5.4) should be undertaken to reduce the radon levels to well below 100
Bq/litre. Appropriate treatments include air stripping, aeration systems or – for small
water supplies – activated carbon adsorption. If there are significant amounts of
radon-producing minerals around the water source, then it may be appropriate for
larger drinking-water supplies to test for radon concentration periodically – for
example, every 5 years.

207
 9. RADIOLOGICAL ASPECTS

9.5.4 Treatment and control methods and technical achievability

Radon, being a gas, is relatively easy to remove by air stripping. Removal efficiencies
of >99% were obtained with diffuse bubble and packed tower aeration at air:water
ratios of 15:1 and 5:1, respectively (Kinner et al., 1990). Other investigations focusing
on aeration at public waterworks have given similar results, with 67–99% efficiencies
(Annanmäki & Turtiainen, 2000). This is the preferred method of treatment.
GAC is also effective in removing radon from water, with removals of 70–100%
(Lykins et al., 1992). The amount of radon removed by activated carbon is effectively
unlimited, because the adsorbed radon decays into other radioactive products, such
as 210Pb. As the adsorbed radon decays, radioactive progeny emitting gamma radia-
tion is produced, possibly creating a disposal problem (Castle, 1988). Elevated gamma
dose rates (up to 120 mSv/h) near the filter have been recorded (Annanmäki & Tur-
tiainen, 2000). Screening of the GAC filter could be required. In some circumstances,
a twin tank system, which introduces a time delay that allows the radon to decay to a
significant extent, may be a low-cost option.

9.6 Sampling, analysis and reporting

9.6.1 Measuring gross alpha and gross beta activity concentrations
To analyse drinking-water for gross alpha and gross beta activities (excluding radon),
the most common approach is to evaporate a known volume of the sample to dryness
and measure the activity of the residue. As alpha radiation is easily absorbed within
a thin layer of solid material, the reliability and sensitivity of the method for alpha
determination may be reduced in samples with a high TDS content.

207a
 GUIDELINES FOR DRINKING-WATER QUALITY

Table 9.4 Methods for the analysis of gross alpha and gross beta activities in drinking-water
Method, reference Technique Detection limit Application
International Organization Evaporation 0.02–0.1 Bq/litre Groundwater with TDS greater
for Standardization: than 0.1 g/litre
ISO-9695 (for gross beta)
ISO-9696 (gross alpha)
(ISO, 1991a, 1991b)
American Public Health Co-precipitation 0.02 Bq/litre Surface water and groundwater
Association (APHA, 1998) (TDS is not a factor)

Where possible, standardized methods should be used to determine concentrations

of gross alpha and gross beta activities. Three procedures for this analysis are listed in
Table 9.4.
The determination of gross beta activity using the evaporation method includes
the contribution from potassium-40. An additional analysis of total potassium is
therefore required if the gross beta screening value is exceeded.
The co-precipitation technique (APHA, 1998) excludes the contribution due to
potassium-40; therefore, determination of total potassium is not necessary. This
method is not applicable to assessment of water samples containing certain fission
products, such as caesium-137. However, under normal circumstances, concentrations
of fission products in drinking-water supplies are extremely low.

9.6.3 Measuring radon

There are difficulties in deriving activity concentrations of radon-222 in drinking-
water arising from the ease with which radon is released from water during handling.
Stirring and transferring water from one container to another will liberate dissolved
radon. According to the widely used Pylon technique (Pylon, 1989, 2003), detection
of radon in drinking-water is performed using a water degassing unit and Lucas scin-
tillation chambers. Water that has been left to stand will have reduced radon activity,
and boiling will remove radon completely.

208
 9. RADIOLOGICAL ASPECTS

9.6.4 Sampling
New groundwater sources for public supplies should be sampled at least once to deter-
mine their suitability for drinking-water supply before design and construction to
characterize the radiological quality of the water supply and to assess any seasonal
variation in radionuclide concentrations. This should include analysis for radon and
radon daughters.
Once measurements indicate the normal range of the supply, then the sampling
frequency can be reduced to, for example, every 5 years. However, if sources of
potential radionuclide contamination exist nearby (e.g., mining activity or nuclear
reactors), then sampling should be more frequent. Less significant surface and under-
ground drinking-water sources may be sampled less frequently.
Levels of radon and radon daughters in groundwater supplies are usually stable
over time. Monitoring of water for radon and its daughters can therefore be relatively
infrequent. Knowledge of the geology of the area should be considered in determin-
ing whether the source is likely to contain significant concentrations of radon and
radon daughters. An additional risk factor would be the presence of mining in the
vicinity; in such circumstances, more frequent monitoring may be appropriate.
Guidance on assessing water quality, sampling techniques and programmes and the
preservation and handling of samples is given in the Australian and New Zealand
Standard (AS, 1998).

9.6.5 Reporting of results

The analytical results for each sample should contain the following information:
— sample identifying code or information;
— reference date and time for the reported results (e.g., sample collection date);
— identification of the standard analytical method used or a brief description of
any non-standard method used;
— identification of the radionuclide(s) or type and total radioactivity determined;
— measurement-based concentration or activity value calculated using the appro-
priate blank for each radionuclide;
— estimates of the counting uncertainty and total projected uncertainty; and
— minimum detectable concentration for each radionuclide or parameter
analysed.
The estimate of total projected uncertainty of the reported result should include
the contributions from all the parameters within the analytical method (i.e., count-
ing and other random and systematic uncertainties or errors).

209